[Bug] Error build in GH200

[johnnynunez]

Checklist

• 1. I have searched related issues but cannot get the expected help.
• 2. The bug has not been fixed in the latest version.
• 3. Please note that if the bug-related issue you submitted lacks corresponding environment info and a minimal reproducible demo, it will be challenging for us to reproduce and resolve the issue, reducing the likelihood of receiving feedback.
• 4. If the issue you raised is not a bug but a question, please raise a discussion at https://github.com/kvcache-ai/ktransformers/discussions. Otherwise, it will be closed.
• 5. To help the community, I will use Chinese/English or attach an Chinese/English translation if using another language. Non-Chinese/English content without translation may be closed.

Describe the bug

If you build from source you get:
get_cpu_instruct index out of range due that I think that it not has compatibility with armv8 and armv9

Reproduction

pip wheel .

Environment

GH200 Arm Neoverse v2 ubuntu 24.04 Cuda 12.8.1

[jizhilong]

get_cpu_instruct index out of range due that I think that it not has compatibility with armv8 and armv9

yes, it is very likely related to the sources' compatibility with cpu processor. A inclusion of the raw traceback and the content of the file /proc/cpuinfo should help the troubleshooting.

[johnnynunez]

Now the error for me is in llamafile:
#1266

but seems that llamafile support arm64

[RWayne93]

you most likely need to build a good amount of the dependencies from source.

[johnnynunez]

you most likely need to build a good amount of the dependencies from source.

I know

[WuNein]

@johnnynunez https://github.com/SakuraTomoya-Wzy/ARM-for-KTransformers

you can refer to this, kt's llamafile missing arm related functions.